Estresse no trabalho e dor lombar crônica - uma análise longitudinal da coorte Elsa-Brasil Musculoesquelético (ELSA- Brasil MSK)

Abstract

INTRODUCTION: Low back pain is the main cause of years lived with disability worldwide and is the most disabling musculoskeletal pain site. It is estimated that approximately 70% of the years lived with disability associated with low back pain occur in people of working age (20-65 years) and it is possible that exposure to work stress may at least partially explain the greater burden of chronic low back pain. (CLBP) in this population group. However, this hypothesis was not verified by a longitudinal study. OBJECTIVES: To investigate the association between stress at work, according to the effort-reward imbalance model (ERI), and the cumulative incidence of CLBP in 4 years of follow-up, considering the number of episodes of pain in the period and the intensity and/or disability of the pain. METHODS: This is a longitudinal study with a three-year follow-up of the cohort of the Longitudinal Study of Adult Health - Musculoskeletal (ELSA-Brasil MSK). Baseline data (2012-2014) and the first three annual telephone interviews were used. A total of 1733 participants without CLBP (lasting >6 months) in the year prior to inclusion in the cohort were included, with information for the ERI scale at baseline of the ELSA-Brasil MSK and for CLBP in the three annual telephone follow-ups. In the follow-ups, CLBP was defined by the presence of low back pain with moderate/severe/very severe intensity and duration >3 months in the last month. The present study had three response variables: 1) Accumulated incidence of CLD (yes/no); 2) Number of CLBP episodes (0, 1, ≥2 episodes); 3) Accumulated incidence of CLD according to intensity/disability (absent/very mild/mild; moderate; severe/very severe and/or disabling). Stress at work was assessed using the long version of the ERI questionnaire. Thus, the explanatory variables of interest were: the effort-reward imbalance, assessed by the effort/reward ratio, and each dimension of the ERI scale (effort, reward and overcommitment) separately. All dimensions were categorized into tertiles of their distributions. We used the causal diagram (Direct Acyclic Graph - DAG) as an auxiliary instrument in defining the variables to be considered in the adjustments of the models that considered prior knowledge about common risk factors for stress at work and low back pain available in the literature. Thus, sociodemographic variables [age (in years); sex; race/color (white/yellow, brown, black/indigenous) and schooling (complete higher education; complete high school; elementary school - complete/incomplete) and occupation characteristics [nature of occupation (non-routine non-manual, non-routine manual and manual) and work shift (only day, night or mixed)]. The estimation for the associations between each of the variables of stress at work and the cumulative incidence of CLBP were made using Poisson regression models. The associations of stress at work variables with the number of CLBP episodes and with the cumulative incidence of CLBP according to intensity/disability were investigated using multinomial logistic regression models, as these are variables with three categories. RESULTS: The cumulative incidence of CLBP in 4 years of follow-up was 24.8%. Lower reward (1st versus 3rd tertiles; RR=1.24; 95%CI=1.01-1.54) and greater overcommitment (3rd versus 1st tertiles; RR=1.23; 95%CI=1.01-1, 50) were associated with a higher incidence of CLBP. Of the total participants, 10.7% reported multiple episodes of CLBP (≥2 episodes). Greater effort (3rd versus 1st tertiles; OR=1.48; 95%CI=1.02-2.15), greater effort/reward ratio (3rd versus 1st tertiles; OR=1.67; 95%CI=1.12- 2.47) and greater overcommitment (3rd versus 1st tertiles; OR=1.67; CI=1.11-2.50) were associated with a greater chance of having multiple CLBP episodes. Severe/disabling CLBP was reported by 9.92% of participants. Lower reward (1st versus 3rd tertiles; OR=1.58; 95%CI=1.02-2.44), higher effort/reward ratio (3rd versus 1st tertiles; OR=1.70; 95%CI=1.14- 2.53), and greater overcommitment (3rd versus 1st tertiles; OR=1.57; 95%CI=0.5-2.34), were independently associated with the cumulative incidence of severe/disabling CLBP. CONCLUSION: The results showed that stress at work, especially greater overcommitment, was associated with a higher risk of CLBP, higher risk of multiple CLBP episodes, and higher risk of severe and/or disabling CLBP after 4 years of follow-up. It is suggested that measures that mitigate exposure to stress at work will possibly reduce the burden of CLBP.